RESEARCH ARTICLE

Adenovirus vector-based multi-epitopevaccine provides partial protection against H5,H7, and H9 avian influenza virusesAhmed O. Hassan1, Omar Amen1,2, Ekramy E. Sayedahmed1, Sai V. Vemula1,Samuel Amoah3, Ian York3, Shivaprakash Gangappa3, Suryaprakash Sambhara3*, SureshK. Mittal1*

1 Department of Comparative Pathobiology and Purdue Institute for Inflammation, Immunology, andInfectious Disease, Purdue University, West Lafayette, IN, United States of America, 2 Poultry DiseasesDepartment, Faculty of Veterinary Medicine, Assiut University, Assiut, Egypt, 3 Influenza Division, Centersfor Disease Control and Prevention, Atlanta, GA, United States of America

* mittal@purdue.edu (SKM); ssambhara@cdc.gov (SS)

Abstract

The emergence of H5, H7, and H9 avian influenza virus subtypes in humans reveals their

pandemic potential. Although human-to-human transmission has been limited, the genetic

reassortment of the avian and human/porcine influenza viruses or mutations in some of the

genes resulting in virus replication in the upper respiratory tract of humans could generate

novel pandemic influenza viruses. Current vaccines do not provide cross protection against

antigenically distinct strains of the H5, H7, and H9 influenza viruses. Therefore, newer vac-

cine approaches are needed to overcome these potential threats. We developed an egg-

independent, adenovirus vector-based, multi-epitope (ME) vaccine approach using the

relatively conserved immunogenic domains of the H5N1 influenza virus [M2 ectodomain

(M2e), hemagglutinin (HA) fusion domain (HFD), T-cell epitope of nucleoprotein (TNP). and

HA α-helix domain (HαD)]. Our ME vaccine induced humoral and cell-mediated immune

responses and caused a significant reduction in the viral loads in the lungs of vaccinated

mice that were challenged with antigenically distinct H5, H7, or H9 avian influenza viruses.

These results suggest that our ME vaccine approach provided broad protection against the

avian influenza viruses. Further improvement of this vaccine will lead to a pre-pandemic

vaccine that may lower morbidity, hinder transmission, and prevent mortality in a pandemic

situation before a strain-matched vaccine becomes available.

Introduction

Human infections caused by H5, H7, and H9 avian influenza virus subtypes provide evidenceof these strains’ pandemic potential [1, 2]. The highly pathogenic avian influenza (HPAI)H5N1 virus initially emerged in China, and subsequently spread to over 60 countries on threecontinents. H5N1 is currently endemic among the poultry populations in South East Asia andAfrica. Over 859 confirmed human infections of H5N1 and 453 deaths (a case fatality rate ofapproximately 52%) have been reported since 2003 [3].

PLOS ONE | https://doi.org/10.1371/journal.pone.0186244 October 12, 2017 1 / 15

a1111111111a1111111111a1111111111a1111111111a1111111111

OPENACCESS

Citation: Hassan AO, Amen O, Sayedahmed EE,Vemula SV, Amoah S, York I, et al. (2017)Adenovirus vector-based multi-epitope vaccineprovides partial protection against H5, H7, and H9avian influenza viruses. PLoS ONE 12(10):e0186244. https://doi.org/10.1371/journal.pone.0186244

Editor: Sang-Moo Kang, Center for Inflammation,Immunity & Infection, Institute for BiomedicalSciences, UNITED STATES

Received: April 21, 2017

Accepted: September 27, 2017

Published: October 12, 2017

Copyright: This is an open access article, free of allcopyright, and may be freely reproduced,distributed, transmitted, modified, built upon, orotherwise used by anyone for any lawful purpose.The work is made available under the CreativeCommons CC0 public domain dedication.

Data Availability Statement: All relevant data arewithin the paper.

Funding: This work was supported by the PublicHealth Service grant AI059374 from the NationalInstitute of Allergy and Infectious Diseases, and theHatch funds. The funders had no role in studydesign, data collection and analysis, decision topublish, or preparation of the manuscript.

In 2013, a new H7N9 avian influenza virus strain emerged in China [4] and has sincecaused more than 1589 human infections and 616 deaths [5]. A H7N9 virus isolate (A/Shang-hai/2/2013) was found to efficiently infect ferrets and pigs with virus shedding, and the viruswas efficiently transmitted from infected ferrets to uninfected ferrets via direct contact [6, 7].Qi et al. suggest that the H7N9 virus has the potential to spread among human populations[8]. However, structural analysis indicates that H7N9 hemagglutinin (HA) characteristics aresimilar to avian HA, including a strong binding affinity to α2-3-linked glycans [9]. These simi-larities offer evidence that current H7N9 viruses are poorly adapted for human-to-human dis-semination. Still, there exists a possibility for H7N9 viruses to acquire the capability to be thenext pandemic virus. Human cases of other H7 influenza viruses (e.g., H7N2, H7N3, andH7N7) linked to poultry outbreaks have been reported in the Netherlands, Canada, Mexico,Italy, the U.S., and the U.K. [10, 11]. Moreover, H9N2 viruses have caused human infectionsin Hong Kong, Bangladesh, and South Korea [10, 12, 13]. Although H5N1, H7N7, H7N9, andH9N2 avian influenza viruses are not readily transmissible among humans, it is possible thatany of these viruses could reassort with a circulating human influenza strain, or acquire therequired mutations of the HA gene segment, and able to attach to human-like α2-6-sialic acidreceptors [14, 15]. This possibility is very likely to result in a novel pandemic influenza strain.

For pandemic preparedness purposes, H5N1, H7N7, and H9N2 virus vaccines have beendeveloped and clinically evaluated. However, the inactivated or subunit H5 vaccines werefound to be low immunogenic, which required the administration of an adjuvant, higherHA doses, or the administration of multiple doses in preclinical and clinical trials [16–18].Furthermore, the diversity of the H5N1 virus has expanded exponentially in the past few years.The virus now has evolved into genetically distinct clades and subclades, which makes thedevelopment of a strain-matched vaccine using traditional methods extremely challenging.The development efforts for H7 and H9 influenza virus vaccines pose similar concerns [19,20]. Recently, a vaccine against H7N1 that was developed in cell culture was proven to bepoorly immunogenic in humans, except when administered with an adjuvant [20]. In addition,a study has shown that the H9N2 vaccine failed to provide cross-reactivity against heterolo-gous viruses from a different clade in human clinical trials, despite the use of an MF59 adju-vant in conjugation with the vaccine [21]. Hence, there is a pressing need for the developmentof a pre-pandemic influenza vaccine that can induce both humoral and cellular immuneresponses and provide broad protection against a wide range of avian influenza viruses cur-rently emerging from avian reservoirs.

Adenovirus (Ad) vector-based vaccines induce excellent humoral and cell-mediatedimmune (CMI) responses [22, 23] due to the adjuvant-like ability of Ad vectors to stimulatethe innate immune system through both Toll-like receptor (TLR)-dependent and TLR-inde-pendent pathways [24, 25]. Ad vector-based influenza vaccines have shown excellent potentialin both animal models [26–28] and clinical trials [29, 30]. Our immunogenicity and protectiveefficacy studies in mice demonstrate that Ad vector-based vaccines provide complete protec-tion against homologous and antigenically distinct influenza virus strains [28].

The M2 ectodomain (M2e) [31–35], HA fusion domain (HFD) [36–39], T-cell epitope ofnucleoprotein [NP] (TNP) [40–42], and HA α-helix domain (HαD) [36–39, 43–47] have beenidentified as relatively conserved immunogenic domains with the potential to provide cross-protection against influenza viruses. These epitopes are either hidden within the native proteinstructure or poorly immunogenic without the use of a strong adjuvant. We hypothesize thatusing these immunogenic domains and their expression in an Ad vector will result in theinduction of optimal immune responses against cross-protective epitopes. The present studydescribes the Ad vector-based, multi-epitope (ME) vaccine approach for broadening protectiveimmunity against the H5, H7, and H9 avian influenza virus subtypes. An Ad vector expressing

Broadly protective avian influenza vaccine

PLOS ONE | https://doi.org/10.1371/journal.pone.0186244 October 12, 2017 2 / 15

Competing interests: The authors have declaredthat no competing interests exist.

ME (M2e, HFD, TNP and HαD) of a H5N1 avian influenza virus was generated and evaluatedfor its immunogenicity and protective efficacy in a mouse model. Immunogenicity studiesdemonstrated that M2e, HFD, TNP, and HαD epitopes were recognized in the Ad vector for-mulation. In order to mimic a pandemic situation, antigenically distinct H5N2, H7N9, andH9N2 avian influenza viruses were used to evaluate our vaccine’s protection efficacy. Theimmunogenicity and protection study results suggest that the development of a pre-pandemicinfluenza vaccine using the relatively conserved immunogenic epitope approach is possible.

Materials and methods

Cells and viruses

293 (human embryonic kidney cells that express E1 gene products of human Ad-C5), 293Cre(293 cell line that expresses Cre recombinase), MDCK (Madin-Darby canine kidney), andBHH2C (bovine-human hybrid clone 2C) [48] cell lines were used for this study. These cellswere grown as monolayer cultures in minimum essential medium (MEM) (Life Technologies,Gaithersburg, MD) supplemented with 10% reconstituted fetal bovine serum (Hyclone,Logan, UT) and gentamycin (50 μg/ml).

The following low pathogenic influenza A viruses were used in this study: A/chukar/MN/14951-7/1998(H5N2), A/goose/Nebraska/17097/2011(H7N9), and A/Hong Kong/1073/1999(H9N2). These influenza viruses were grown in embryonated hen eggs and were titrated ineggs, MDCK, or both. These viruses will not cause significant morbidity or mortality in micesince they were not mouse-adapted. Therefore, the lung virus titers following challenge will bethe best measurable parameter to evaluate the vaccine’s efficacy.

Generation of human Ad replication defective vectors

Human Ad vectors were generated in 293Cre cells using Cre-recombinase site-specific recom-bination [49]. The construction of Ad-H5HA, Ad-H7HA, and Ad-H9HA, each containing afull-length coding region of H5 HA of A/Vietnam/1203/2004(H5N1), H7 HA of A/Nether-lands/219/2003(H7N7), or H9 HA of A/chicken/Hong Kong/G9/1997(H9N2), respectively aspreviously described [50, 51]. The polybasic cleavage sites of full HA genes of H5 and H7 weremodified to match the low pathogenic strains. The generation of empty vector Ad-ΔE1E3 (E1& E3 deleted Ad-C5) is described elsewhere [52].

For the ME vaccine, the following M2e, HFD, TNP, and HαD domains of H5N1 (A/Viet-nam/1203/2004) were separated by the AlaAlaAla, GlyGlyGly and GlyProGlyProGly linkers,respectively:

M2e: MSLLTEVETPTRNEWECRCSDSSD;

HFD: GLFGAIAGFIEGGW;

TNP: TYQRTRALVR;

HαD: RIENLNKKMEDGFLDVWTYNAELLVLMENERTLDFHDSNVKNLYDKVRLQLRD

NA.

The H5N1 ME (H5ME) gene cassette was synthesized commercially. The construct struc-tures and their names are shown in Fig 1. The Ad vector containing H5ME was generated in293Cre cells using Cre-recombinase site-specific recombination. The gene construct wasunder the control of the human cytomegalovirus (HCMV) immediate early promotor and thebovine growth hormone (BGH) polyadenylation signal. All Ad constructs were purified usingcesium chloride density-gradient ultracentrifugation [53] and titrated in BHH2C cells [48].

Broadly protective avian influenza vaccine

PLOS ONE | https://doi.org/10.1371/journal.pone.0186244 October 12, 2017 3 / 15

Animal inoculation and protection studies

All animal experiments were conducted in a USDA-approved BSL-2+ facility at Purdue Uni-versity with the approval and in accordance with the guidelines of the Institutional BiosafetyCommittee and Institutional Animal Care and Use Committee. Six-to-eight-week-old femaleBALB/c mice (purchased from Harlan Sprague Dawley Inc., Indianapolis) were used to evalu-ate the breadth of immune responses and protection efficacy. Immunization and challengeexperiments were conducted in two phases. During the first phase, five groups of mice (7 ani-mals/group) were immunized via intramuscular inoculation of 108 PFU of each of Ad vectorsas follows: Ad-H5HA, Ad-H7HA, Ad-H9HA, Ad-H5ME, or empty vector Ad-ΔE1E3. Theimmunization was administered twice within a three-week interval. The animals were anesthe-tized with ketamine-xylazine (90 mg/kg ketamine and 10 mg/kg xylazine) via intraperitonealinjections three weeks after the second immunization. Blood samples were collected via retro-orbital puncture to evaluate the humoral immune responses, and the spleens were collectedupon euthanasia to evaluate the cell-mediated immune responses. The mice under anesthesiawere euthanized using cervical dislocation and the thorax was perforated to ensure death.These procedures were in accordance with the recommendations of the American VeterinaryMedical Association’s Panel on Euthanasia.

During the second phase, similar vaccinations as described for the first phase were followedwith five animals in each group. Three weeks after the final vaccination, the animals were chal-lenged intra-nasally with a 100 mouse infectious dose 50 (MID50) of A/chukar/MN/14951-7/1998(H5N2), A/goose/Nebraska/17097/2011(H7N9) or A/Hong Kong/1073/1999(H9N2).Three days after the challenge, the animals were euthanized under anesthesia with ketamine-xylazine (90 mg/kg ketamine and 10 mg/kg xylazine) via intraperitoneal injections, and thelungs were collected and then homogenized in 1 ml of sterile phosphate-buffered saline (PBS).A ten-fold serial dilution of the lung homogenates was used to infect either MDCK cells in96-well plates (for H5N2 and H9N2 virus-challenged groups) or specific-pathogen free (SPF)embryonated hen eggs (for H7N9 virus-challenged groups). Following 72 h incubation at37˚C, the infected-cell supernatants or allantoic fluid samples were tested by hemagglutinationassay (HA) using 0.5% turkey red blood cells (TRBC) to determine a 50% tissue culture infec-tious dose (TCID50) [for H5N2 and H9N2 virus-challenged groups] or 50% egg infectiousdose (EID50) [for H7N9 virus-challenged groups] using the Reed-Muench method [54].

Hemagglutination inhibition assay

The mouse serum samples were treated with receptor-destroying enzyme (RDE) [Denka Sei-ken, Tokyo, Japan] and incubated at 37˚C for 20 h. Thereafter the samples were inactivated by

Fig 1. Schematic diagram of adenovirus (Ad) vector constructs used in the study. (A) Ad-ΔE1E3;human adenovirus type C5 (Ad-C5) empty vector. (B) Ad-H5HA, Ad-H7HA and Ad-H9HA; Ad-C5 vectorcontaining a full-length coding region of H5 hemagglutinin (HA) of A/Vietnam/1203/04(H5N1), H7 HA of A/Netherlands/219/2003(H7N7) or H9 HA of A/chicken/Hong Kong/G9/1997(H9N2), respectively. (C) Ad-H5ME, Ad-C5 vector containing multi-epitope (ME) gene construct of A/Vietnam/1203/ 2004(H5N1) thatcontains the extracellular domain of M2 (M2e), fusion domain of HA (HFD), T-cell epitope of nucleoprotein[NP] (TNP), and alpha helix domain of HA (HαD).

https://doi.org/10.1371/journal.pone.0186244.g001

Broadly protective avian influenza vaccine

PLOS ONE | https://doi.org/10.1371/journal.pone.0186244 October 12, 2017 4 / 15

incubation at 56˚C for 30 min. Two-fold dilutions of RDE-treated sera were conducted in96-well plates first with a 1:10 dilution and followed by four hemagglutination units (HAU) ofeach virus diluted in PBS. Following incubation at room temperature for 1 h, 0.5% TRBC wasadded and the plates were incubated at room temperature for 20 min to allow hemagglutina-tion to occur. The hemagglutination inhibition (HI) antibody titer was determined as thereciprocal of the highest dilution that completely prevented TRBC from agglutination, and thedata are presented as the geometric mean HI titers.

Virus neutralization assays

The assays were performed as previously described [28]. Two-fold serial dilutions of heat-inac-tivated serum samples were mixed with 100 TCID50 or 100 EID50 of an influenza virus, andthe mixture was incubated for 1 h at 37˚C. The mixture was then added to MDCK cells in96-well plates or injected into the allantoic fluid of embryonated hen eggs and incubated for 72h at 37˚C. Thereafter, the HA activities of the cell supernatants or allantoic fluid samples weredetected by HA assay with 0.5% TRBC. The virus neutralization (VN) titer was determined asthe reciprocal of the highest dilution that completely neutralized the virus, and the data arepresented as the geometric mean VN titers.

Enzyme-linked immunosorbent assay (ELISA)

ELISA was performed as described earlier [50, 55]. The M2e, HαD, and HFD peptides biotiny-lated at the N-terminus were commercially synthesized (Gen Script, Piscataway, NJ). 96-wellELISA plates (eBioscience, San Diego, CA) were coated with streptavidin and incubated over-night at 37˚C. Then the biotinylated M2e, HαD, or HFD peptide (5 μg/ml) were added andincubated at room temperature for 1 h. After blocking the non-specific protein-binding siteswith PBS containing 1% bovine serum albumin (BSA), diluted serum samples (1:50 dilutionfor M2e or HFD and 1:200 dilution for HαD) were added and incubated at room temperaturefor 3 h. During the standardization process, various dilutions of the serum samples for eachpeptide were used to determine the best serum dilution for each peptide. The horseradish per-oxidase-conjugated goat anti-mouse IgG (Sigma-Aldrich, Inc., St. Louis, MO) at a dilution of1:2000 was added to the wells, followed by 1 h incubation at room temperature. The plateswere developed with a TMB substrate (Biolegend, San Diego, CA). The reaction was stoppedby adding 1 M sulfuric acid, and the optical density measurements were taken at 450 nm usinga microplate reader (Molecular Devices, Sunnyvale, CA).

ELISpot assay

The assay was performed as previously described [28]. The spleens collected from mice fol-lowing three weeks post-booster immunization were homogenized using cell strainers (FisherScientific, Hampton, NH). The red blood cells were lysed with the ACK lysis buffer (LonzaWalkersville, Inc., MD). The splenocytes were then resuspended in RPMI-1640 medium con-taining L-glutamine and 25mM HEPES (Gibco, Grand Island, NY), 10% reconstituted fetalbovine serum, nonessential amino acids, sodium pyruvate, β-mercaptoethanol, penicillin/streptomycin, and gentamycin. The viable cell count was determined by mixing splenocyteswith 0.4% Trypan Blue stain (Fisher Scientific, Hampton, NH) and then counting the unstainedlive cells on a hemocytometer under a light microscope.

PVDF-backed 96-well microplates (Millipore, Billerica, MA) were coated overnight at 4˚Cwith a monoclonal antibody for mouse interferon-gamma (INFγ) [BD Bioscience, San Jose,CA]. 106 splenocytes (in RPMI-1640 medium with supplements) were incubated with eitherHA518 (IYSTVASSL) or NP147 (TYQRTRALV) peptides (H-2Kd-restricted cytotoxic T

Broadly protective avian influenza vaccine

PLOS ONE | https://doi.org/10.1371/journal.pone.0186244 October 12, 2017 5 / 15

lymphocyte (CTL) epitopes for HA and NP, respectively) for 60 h at 37˚C in a CO2 incubator.As a positive control, the splenocytes were cultured in the presence of phorbol myristate ace-tate/ionomycin (Sigma-Aldrich, St. Louis, MO). The cells were removed by washing, and abiotinylated rat anti-mouse antibody specific for INFγ (BD Bioscience, San Jose, CA) wasadded. The plates were incubated overnight at 4˚C. Following the addition of streptavidin-AP(Sigma-Aldrich, St. Louis, MO) and washing, a chromogen mixture of BCIP/NBT (Sigma-Aldrich, St. Louis, MO) was added for color development. The reaction was stopped by rinsingthe plates with deionized water. The ELISpot plates were read using a Bioreader 5000 (BIOSYS,Miami, FL) to count the number of spot-forming units (SFU).

Statistical analyses

For statistical significance determination, one- and two-way ANOVA tests were conductedwith Dunnett’s post-hoc analysis at a significance level of p<0.05.

Results

Generation of Ad vector expressing ME of H5N1 influenza virus

A Cre-recombinase site-specific recombination system was used to insert a coding sequence ofthe ME containing M2e, HFD, TNP, and HαD of H5N1 (H5ME) into the E1 region of Ad-C5to generate Ad-H5ME (Fig 1). The recombinant virus showed cytopathic effects 7–10 dayspost-transfection of 293Cre cells. The construction and characterization of Ad-H5HA, Ad-H7HA, and Ad-H9HA have been previously described [50, 51].

Induction of humoral immune responses against H5ME

The mice were inoculated i.m. with various Ad vectors, as described in the Material and Meth-ods section. Three weeks following the booster inoculation, serum samples were collected andassayed for M2e-, HαD-, or HFD-specific IgG by ELISA (Fig 2). In this study, we did not mon-itor the immunogenicity of the prime dose (single inoculation prior to the second dose of thevaccine); however, it seems that the immunization of mice with a single dose of Ad vector-based vaccine elicited protective immune responses, and the second inoculation resulted inmeasurable increases in immunogen-specific immune responses [56].

Fig 2. Influenza antigen-specific antibody responses generated after immunization with an Ad vector-based multi-epitope (ME) vaccine. Threeweeks after the second inoculation, serum samples were collected from all animal groups as described in the material and methods section, and tested byELISA for IgG antibody responses specific to M2e (A), HαD (B) or HFD (C). Data are represented as the mean ± standard deviation (SD) of the optical density(OD) readings. Based on OD values, M2e-, HαD- and HFD-specific IgG responses induced by H5ME were statistically significant (*P 0.05 as compared tothe ΔE1E3 control). H5HA, Ad-H5HA; H7HA, Ad-H7HA; H9HA, Ad-H9HA, H5ME, Ad-H5ME; ΔE1E3, Ad-ΔE1E3.

https://doi.org/10.1371/journal.pone.0186244.g002

Broadly protective avian influenza vaccine

PLOS ONE | https://doi.org/10.1371/journal.pone.0186244 October 12, 2017 6 / 15

Mice vaccinated with Ad vector expressing H5ME (Ad-H5ME) showed high levels of M2e-specific IgG (Fig 2A). For HαD-specific IgG responses, mice vaccinated with Ad-H5ME aswell as Ad-H5HA (vector containing the full H5HA gene) showed elevated levels of HαD-spe-cific IgG (Fig 2B). Similarly, HFD-specific IgG responses increased in mice vaccinated withAd-H5ME (Fig 2C).

The serum samples from immunized mice were also analyzed for HI titers to determine thepresence of HA-specific antibodies against heterologous H5 (H5N2), H7 (H7N9), and H9(H9N2) avian influenza viruses, as we have previously elucidated that Ad vectors elicit en-hanced HI titers against homologous avian influenza viruses [50]. HI titers (320) against theH5N2 virus were observed in the Ad-H5HA-immunized group only (Table 1). Detectable HIactivities (176) against the H7N9 virus were noticed in the Ad-H7HA-inoculated group only(Table 1). Similarly, HI antibody titers (44) against the H9N2 virus were detected in the Ad-H9HA-inoculated group only (Table 1).

The serum samples from various inoculated groups were also analyzed for neutralizingantibodies against the H5N2, H7N9, and H9N2 influenza viruses. Since the influenza virusesused for microneutralization assays have antigenically distinct HAs, lower levels of virus neu-tralizing antibody titers were anticipated depending on the differences between the H5, H7,and H9 HA sequences in the Ad vectors, and the HA sequences of the H5N2, H7N9, andH9N2 influenza viruses used for microneutralization assays. Mice vaccinated with Ad vectorscontaining the HA construct, Ad-H5HA, Ad-H7HA, or Ad-H9HA developed 246, 780, and 31geometric mean VN titers against H5N2, H7N9, and H9N2 influenza viruses, respectively(Table 1). Noticeably, the Ad-H5ME-inoculated group did not develop HI or VN antibodiesagainst H5N2, H7N9, or H9N2 viruses (Table 1).

Induction of cell-mediated immune responses against H5ME

It is well known that CMI responses against influenza viruses play a role in virus clearance[57]. To evaluate the CMI responses against influenza in immunized mice, the spleens fromAd-vector inoculated groups were collected three weeks after the second immunization. Thesplenocytes were isolated and analyzed for HA- or NP-specific CMI responses following invitro stimulation with HA518 or NP147 peptide, respectively, using an IFNγ-specific ELISpotassay. Due to the fact that HA518 epitope, a CD8 T-cell epitope, is conserved in H5HA andH9HA, but not in H7HA, the animal groups that received Ad-H5HA or Ad-H9HA showedhigher numbers of INFγ-secreting HA518-specific CD8 T cells compared to the empty vector

Table 1. Hemagglutination inhibition (HI) and virus neutralization (VN) antibody titers in mice vaccinated with an Ad vector-based multi-epitope(ME) vaccine.

Vaccine group H5N2 H7N9 H9N2

HI VN HI VN HI VN

ΔE1E3 <10 <10 <10 <10 <10 <10

H5HA 320 246 <10 <10 <10 <10

H7HA <10 <10 176 780 <10 <10

H9HA <10 <10 <10 <10 44 31

H5ME <10 <10 <10 <10 <10 <10

Mice (7 mice/group) were immunized twice with H5HA, H7HA, H9HA, H5ME or ΔE1E3 at an interval of three weeks. Three weeks after the final

immunization, serum samples were collected for detecting HI and VN antibody titers. The titers are shown as the geometric mean titers. H5HA, Ad-H5HA;

H7HA, Ad-H7HA; H9HA, Ad-H9HA,; H5ME, Ad-H5ME; ΔE1E3, Ad-ΔE1E3; H5N2, A/chukar/MN/145917/1998(H5N2); H7N9, A/goose/Nebraska/17097/

2011(H7N9); H9N2, A/Hong Kong/1073/ 1999(H9N2).

https://doi.org/10.1371/journal.pone.0186244.t001

Broadly protective avian influenza vaccine

PLOS ONE | https://doi.org/10.1371/journal.pone.0186244 October 12, 2017 7 / 15

(Ad-ΔE1E3) control group (Fig 3A). These results are consistent with our previous findings[50]. Mice that were vaccinated with Ad-H5ME showed increased numbers of NP147-specificIFNγ-secreting CD8 T cells compared to the empty vector (Ad-ΔE1E3) control group (Fig 3B).

Breadth of protection conferred by Ad vector-based ME vaccinefollowing challenge with antigenically distinct H5N2, H7N9, or H9N2influenza viruses

One of the important vaccine efficacy measures is the lung viral titers in vaccinated animalschallenged with an influenza virus [28]. A significant decrease in lung viral titers is indicativeof the vaccine protective efficacy and the reduction in the virus transmission. Ad vector-basedvaccine efficacy in mice was evaluated by measuring the lung viral titers three days afterchallenges with a 100 MID50 of H5N2, H7N9, or H9N2 influenza virus. As expected, immuni-zation with Ad-H5HA, Ad-H7HA, or Ad-H9HA resulted in complete protection with ahomosubtypic virus only (Table 2). Animal groups that were vaccinated with Ad-H5MEshowed 1.5, 1.2, and 1.8 log reductions in the lung viral titers of antigenically distinct H5N2,H7N9, or H9N2 influenza virus, respectively, compared to the Ad-ΔE1E3-inoculated groups(Table 2).

Discussion

Incidences of human infections with H5N1, H7N7, H7N1, H7N3, H7N9, and H9N2 avianinfluenza virus subtypes have demonstrated their pandemic potential; however, the character-istics of a new pandemic influenza virus will be unknown until the time of the pandemic. Thedevelopment of a pre-pandemic influenza vaccine seems to be one of the most effective

Fig 3. HA518 and NP147 epitope-specific IFNγ secreting CD8+ T cells in the spleens of vaccinated mice. Three weeks after the second inoculation, thespleens were collected from all animal groups after euthanizing the animals as described in the material and methods section. The splenocytes wereevaluated for HA-specific (A) or NP-Specific (B) cell-mediated immune responses using INFγ-ELISpot assay. The data represent mean ± standard deviation(SD). The number of spot-forming units (SFU) induced by H5ME in NP-specific INFγ-ELISpot, and the number of SFU induced by H5HA or H9HA in HA-specific INFγ-ELISpot were found to be statistically significant (*P 0.05 as compared to the ΔE1E3 control). H5HA, Ad-H5HA; H7HA, Ad-H7HA; H9HA, Ad-H9HA; H5ME, Ad-H5ME; ΔE1E3, Ad-ΔE1E3.

https://doi.org/10.1371/journal.pone.0186244.g003

Broadly protective avian influenza vaccine

PLOS ONE | https://doi.org/10.1371/journal.pone.0186244 October 12, 2017 8 / 15

measures of pandemic preparedness. The pre-pandemic influenza vaccine needs to providebroad cross-protection, given that it is almost impossible to predict if and when H5N1, H7N9,H9N2, or another influenza virus will acquire the characteristics of a pandemic virus. In thismanuscript, our efforts were directed towards the generation of a broadly protective vaccineagainst H5, H7, and H9 influenza subtypes, with the anticipation that such a vaccine wouldsignificantly lower morbidity, hinder transmission, and prevent mortality in a pandemic situa-tion before a strain-matched vaccine can be produced. Our vaccine could also prime humanpopulations for improved immune responses in the aftermath of pandemic virus infections.

The relatively conserved influenza A virus epitopes have been investigated for their poten-tial as immunogens to induce cross-protective immune responses. Vaccine strategies targetingrelatively conserved domains, for example M2e and HA2, have the ability to induce cross-reac-tive humoral immune responses [36, 58]. Internal proteins containing relatively conserved T-cell epitopes (e.g., M1, PB1, and NP) were also investigated for their ability to induce cross-protective T-cell responses [59, 60]. In the present study, we attempted to combine the hetero-subtypic CMI response against TNP and the cross-reactive (not necessarily cross-neutralizing)humoral immune responses against M2e, HFD, and HαD domains to provide broad protec-tion against a potential H5N1, H7N9, or H9N2 avian influenza virus pandemic. The Ad vectorsystem was used to deliver ME containing the relatively conserved B- and T-cell epitopes(M2e, HFD, TNP, and HαD) of H5N1 in a single vector formulation. To mimic an influenzapandemic, we purposely used three antigenically distinct wild type influenza viruses, A/chu-kar/MN/14951-7/1998(H5N2), A/goose/Nebraska/17097/2011(H7N9) and A/Hong Kong/1073/1999(H9N2), to represent the H5N1, H7N7, and H9N2 viruses, respectively.

In the present study, the Ad-H5ME expressing ME of H5N1 (H5ME) was recognized bythe host immune system as evident with significant levels of M2e-, HFD-, and HαD-specificantibody responses as well as the T-cell response to the TNP epitope. As expected, Ad-H5MEdid not induce H5N2-specific HI or VN antibodies in immunized animals. However, follow-ing challenges with H5N2, H7N9, or H9N2, there was approximately 1.5, 1.2, and 1.8 logreductions in lung virus titers, respectively, which is indicative of cross-protection. The levelsof cross-reactive humoral immune responses could vary significantly depending on the HAsubtype antigenic group (H5 and H9 in Group 1 and H7 in Group 2) [61]. The lower levels ofcross protection against H7N9 compared to H5N2 or H9N2 in this study may be due to theantigenic differences between Groups 1 and 2.

Table 2. Lung viral titers in vaccinated mice after challenge with H5N2, H7N9 or H9N2 influenza viruses.

Vaccine group Lung virus titer after H5N2 challenge,Log10 TCID50/ml

Lung virus titer after H7N9 challenge,Log10 EID50/ml

Lung virus titer after H9N2 challenge,Log10 TCID50/ml

ΔE1E3 3.9±0.57 4.9±0.27 6.2±0.5

H5HA < 1.2* 4.5±0.27 5.8±0.62

H7HA 3.5+0.35 <1.2* 5.9±0.35

H9HA 3.7±0.43 4.7±0.54 < 1.2*H5ME 2.4±0.32* 3.7±0.54* 4.4±0.6*

Mice (5 mice/group) were immunized twice with H5HA, H7HA, H9HA, H5ME, or ΔE1E3 at an interval of three weeks. Three weeks after the final

immunization, mice were challenged with one of the following viruses: H5N2, H7N9 or H9N2. Three days after the challenged, mice were euthanized and

the lungs were harvested to determine lung virus titers. The data are shown as mean Log10 TCID50±SD (for H9N2 and H5N2) or mean Log10 EID50±SD (for

H7N9). The detection limit of lung viral titer was <1.2 Log10 TCID50/ml (for H9N2 and H5N2) or <1.2 Log10 EID50/ml (for H7N9) [indicated as <1.2].

*P 0.05 as compared to the ΔE1E3 control. H5HA, Ad-H5HA; H7HA, Ad-H7HA; H9HA, Ad-H9HA,; H5ME, Ad-H5ME; ΔE1E3, Ad-ΔE1E3; TCID50, tissue

culture infectious dose 50; EID50, egg infectious dose 50; SD, standard deviation; H5N2, A/chukar/MN/145917/1998(H5N2); H7N9, A/goose/Nebraska/

17097/2011(H7N9); H9N2, A/Hong Kong/1073/ 1999(H9N2).

https://doi.org/10.1371/journal.pone.0186244.t002

Broadly protective avian influenza vaccine

PLOS ONE | https://doi.org/10.1371/journal.pone.0186244 October 12, 2017 9 / 15

The N terminal fusion domain of HA2 (HFD) is considered to be conserved among differ-ent influenza subtypes [62]. HFD was described for their immunological potential and abilityto induce antibodies [63]. Antibodies directed against HFD could inhibit the fusion betweenthe viral and endosomal membrane and neutralize virus infectivity; however, no effective pro-tection was reported exclusively due to HFD. It has been shown that HFD conjugated to key-hole limpet hemocyanin (KLH) has the ability to induce antibodies in mice following twoimmunizations and provide cross protection against homologous and heterologous influenzavirus challenges [64]. On the other hand, HFD is small in size and was found to be less immu-nogenic than M2e [65]. Moreover, the alpha helix domain of HA2 (HαD), when tested asimmunogen, induced antibodies with broader neutralizing activity and provided protectionagainst influenza viruses of distinct subtypes [39, 43, 66]. The inclusion of HFD and HαD epi-topes of the H5, H7, and H9 influenza viruses may further enhance the protection efficacy ofME-based vaccines.

In the present study, the immunization of mice with Ad-based ME vaccine failed to inducestrong virus-neutralizing antibody responses. The Ad-H5ME vaccine failed to induce steriliz-ing immunity against high dose virus challenges, but our vaccine significantly reduced thevirus replication in the lungs of mice challenged with antigenically distinct H5N2, H7N9, andH9N2 influenza viruses. The presence of heterosubtypic protection in the absence of strongneutralizing antibodies indicates the role of other players in providing cross-protection,including CMI responses and non-neutralizing humoral responses against M2e, HFD, andHαD. These findings are consistent with our previous results where the vaccination of micewith Ad-based vaccines expressing H1 HA induced high levels of broadly reactive, stalk-spe-cific antibodies and reduced the lung viral titers of H5N1, H7N2, and H9N2 after challenges[50]. Similarly, Epstein et al. found that a DNA prime-Ad boost with the H1N1 NP inducedCMI responses that protected mice against heterosubtypic challenges [67]. Two other studiesof H3 HA Ad-based vaccine efficacy in mice and pigs have shown that protection against het-erosubtypic challenges does not require the presence of neutralizing antibodies [29, 68].

It is known that HA518-specific CTL are induced in BALB/c mice following influenza virusinfection, but these responses are less pronounced than NP147-specific CTL responses [69].The role of HA518 epitope-specific CD8 T cells in heterologous or heterosubtypic protectionis not clearly understood; however, our protection data from the mouse groups inoculatedwith Ad-H5HA, Ad-H7HA, or Ad-H9HA clearly suggest that HA518-specific CTL did notplay a significant role in heterosubtypic protection.

NP is relatively conserved and represents a target for CTL-mediated immune responsescritical to virus clearance and recovery after infection [40]. Hence, these CMI responses arecross-protective between different influenza A subtypes [70]. The TNP epitope included in ourAd-H5ME vaccine induced NP147 epitope-specific CD8 T cells secreting IFNγ in the vacci-nated group. It is anticipated that NP147 epitope-specific CD8 T cells will contribute to thereduction of lung viral titers after virus challenges. This is consistent with the finding that theinclusion of TNP (NP147-155) with H5 HA in a DNA vaccine resulted in reducing the lungviral titers following a H5N1 challenge [71]. Additional studies have demonstrated that TNPwas found to be immunogenic in mice, induced CD8 T-cell responses, and played a role invirus clearance [72, 73]. In our preliminary work we observed that mice immunized twice with108 PFU of an Ad vector expressing only the TNP resulted in approximately 0.4 log decreasesin virus lung titers following challenge with an influenza virus compared to the animal groupinoculated with the empty Ad vector [Ad-ΔE1E3]. These results suggest that in our ME vaccinestudy, humoral immune responses against M2e, HFD, and HαD domains contributed toapproximately 1.1, 0.8, and 1.4 log reductions in lung virus titers against the H5, H7, and H9

Broadly protective avian influenza vaccine

PLOS ONE | https://doi.org/10.1371/journal.pone.0186244 October 12, 2017 10 / 15

influenza viruses, respectively. Additional studies will be required to evaluate the roles of indi-vidual epitopes in protection against the H5, H7, and H9 influenza viruses.

Pre-existing vector immunity is considered to impact vector uptake and to reduce the leveland duration of transgene expression. We previously demonstrated that moderate levels ofpre-existing vector immunity (<520 virus neutralizing titer) did not significantly hamper theprotective efficacy of an Ad vector-based influenza vaccine in a mouse model [53]. Whereas,the higher levels of pre-existing vector immunity (1500 virus neutralizing titer) could be over-come by either varying the route of immunization or increasing the vaccine dose. Further-more, several other approaches have been developed to circumvent the pre-existing vectorimmunity, including the use of other human Ad serotypes, non-human Ad, and DNA primefollowed by an Ad boost. The impact of vector immunity on the immunogenicity and protec-tive efficacy of Ad-H5ME vaccine should be measured in a separate study.

Future developments toward a universal influenza vaccine may include relatively conservedB cell domains comprised of HA2 and M2e, and conserved T cell domains of NP. Furtherstudies will be required to evaluate the best strategy to allow the improved presentation andrecognition of the epitopes on HA2 and M2e for the development of strong, cross-protectivehumoral immune responses. Additional experiments are necessary to determine an approachfor the vaccine formulation containing NP, HA2, and M2e, which would result in balancedhumoral and CMI responses as well as enhanced cross-protection against different influenzasubtypes.

Acknowledgments

This work was supported by the Public Health Service grant AI059374 from the National Insti-tute of Allergy and Infectious Diseases, and the Hatch funds. The funders had no role in studydesign, data collection and analysis, decision to publish, or preparation of the manuscript. Wethank Jane Kovach, Donna Schrader, GuangJun Zhang, and Sara Penka for their assistance inpreparing the manuscript.

Disclaimer: The findings and conclusions in this report are those of the authors and donot necessarily represent the views of the Centers for Disease Control and Prevention, U.S.Department of Health and Human Services.

Author Contributions

Conceptualization: Suryaprakash Sambhara, Suresh K. Mittal.

Data curation: Ahmed O. Hassan, Omar Amen, Ekramy E. Sayedahmed, Sai V. Vemula, Sam-uel Amoah.

Formal analysis: Ahmed O. Hassan, Ekramy E. Sayedahmed, Suryaprakash Sambhara, SureshK. Mittal.

Funding acquisition: Ian York, Suryaprakash Sambhara, Suresh K. Mittal.

Investigation: Ahmed O. Hassan, Omar Amen, Ekramy E. Sayedahmed, Sai V. Vemula, Sam-uel Amoah, Ian York, Shivaprakash Gangappa, Suryaprakash Sambhara, Suresh K. Mittal.

Methodology: Ahmed O. Hassan, Omar Amen, Ekramy E. Sayedahmed, Sai V. Vemula, Sam-uel Amoah.

Project administration: Ian York, Shivaprakash Gangappa, Suryaprakash Sambhara, SureshK. Mittal.

Resources: Ian York, Shivaprakash Gangappa, Suryaprakash Sambhara, Suresh K. Mittal.

Broadly protective avian influenza vaccine

PLOS ONE | https://doi.org/10.1371/journal.pone.0186244 October 12, 2017 11 / 15

Supervision: Ian York, Shivaprakash Gangappa, Suryaprakash Sambhara, Suresh K. Mittal.

Validation: Ahmed O. Hassan, Omar Amen, Ekramy E. Sayedahmed, Sai V. Vemula, SamuelAmoah, Ian York, Shivaprakash Gangappa, Suryaprakash Sambhara, Suresh K. Mittal.

Writing – original draft: Ahmed O. Hassan, Suryaprakash Sambhara, Suresh K. Mittal.

Writing – review & editing: Ahmed O. Hassan, Omar Amen, Ekramy E. Sayedahmed, Sai V.Vemula, Samuel Amoah, Ian York, Shivaprakash Gangappa, Suryaprakash Sambhara, Sur-esh K. Mittal.

References1. Belser JA, Bridges CB, Katz JM, Tumpey TM. Past, present, and possible future human infection with

influenza virus A subtype H7. Emerg Infect Dis. 2009; 15(6):859–65. Epub 2009/06/16. https://doi.org/10.3201/eid1506.090072 PMID: 19523282.

2. Katz JM, Veguilla V, Belser JA, Maines TR, Van Hoeven N, Pappas C, et al. The public health impact ofavian influenza viruses. Poult Sci. 2009; 88(4):872–9. Epub 2009/03/12. https://doi.org/10.3382/ps.2008-00465 PMID: 19276438.

3. WHO | Cumulative number of confirmed human cases of avian influenza A(H5N1) reported to WHO:World Health Organization; 2017 [updated 2017-08-09 13:07:55]. Available from: http://www.who.int/influenza/human_animal_interface/H5N1_cumulative_table_archives/en/.

4. Gao R, Cao B, Hu Y, Feng Z, Wang D, Hu W, et al. Human infection with a novel avian-origin influenzaA (H7N9) virus. N Engl J Med. 2013; 368(20):1888–97. Epub 2013/04/13. https://doi.org/10.1056/NEJMoa1304459 PMID: 23577628.

5. FAO H7N9 situation update—Avian Influenza A(H7N9) virus—FAO Emergency Prevention System forAnimal Health (EMPRES-AH) 2017. Available from: http://www.fao.org/ag/againfo/programmes/en/empres/h7n9/situation_update.html.

6. Richard M, Schrauwen EJ, de Graaf M, Bestebroer TM, Spronken MI, van Boheemen S, et al. Limitedairborne transmission of H7N9 influenza A virus between ferrets. Nature. 2013; 501(7468):560–3. Epub2013/08/09. https://doi.org/10.1038/nature12476 PMID: 23925116.

7. Zhu H, Wang D, Kelvin DJ, Li L, Zheng Z, Yoon SW, et al. Infectivity, transmission, and pathology ofhuman-isolated H7N9 influenza virus in ferrets and pigs. Science. 2013; 341(6142):183–6. Epub 2013/05/25. https://doi.org/10.1126/science.1239844 PMID: 23704376.

8. Qi X, Qian YH, Bao CJ, Guo XL, Cui LB, Tang FY, et al. Probable person to person transmission ofnovel avian influenza A (H7N9) virus in Eastern China, 2013: epidemiological investigation. BMJ. 2013;347:f4752. Epub 2013/08/08. https://doi.org/10.1136/bmj.f4752 PMID: 23920350.

9. Xu R, de Vries RP, Zhu X, Nycholat CM, McBride R, Yu W, et al. Preferential recognition of avian-likereceptors in human influenza A H7N9 viruses. Science. 2013; 342(6163):1230–5. Epub 2013/12/07.https://doi.org/10.1126/science.1243761 PMID: 24311689.

10. WHO | Avian and other zoonotic influenza: World Health Organization; 2017 [updated 2017-09-0813:05:31]. Available from: http://www.who.int/influenza/human_animal_interface/en/.

11. Fouchier RA, Schneeberger PM, Rozendaal FW, Broekman JM, Kemink SA, Munster V, et al. Avianinfluenza A virus (H7N7) associated with human conjunctivitis and a fatal case of acute respiratory dis-tress syndrome. Proc Natl Acad Sci U S A. 2004; 101(5):1356–61. Epub 2004/01/28. https://doi.org/10.1073/pnas.0308352100 PMID: 14745020.

12. Peiris M, Yuen KY, Leung CW, Chan KH, Ip PL, Lai RW, et al. Human infection with influenza H9N2.Lancet. 1999; 354(9182):916–7. PMID: 10489954.

13. Guo Y, Li J, Cheng X. [Discovery of men infected by avian influenza A (H9N2) virus]. Zhonghua Shi YanHe Lin Chuang Bing Du Xue Za Zhi. 1999; 13(2):105–8. Epub 2003/02/07. PMID: 12569771.

14. Herfst S, Schrauwen EJ, Linster M, Chutinimitkul S, de Wit E, Munster VJ, et al. Airborne transmissionof influenza A/H5N1 virus between ferrets. Science. 2012; 336(6088):1534–41. Epub 2012/06/23.https://doi.org/10.1126/science.1213362 PMID: 22723413.

15. Imai M, Watanabe T, Hatta M, Das SC, Ozawa M, Shinya K, et al. Experimental adaptation of an influ-enza H5 HA confers respiratory droplet transmission to a reassortant H5 HA/H1N1 virus in ferrets.Nature. 2012; 486(7403):420–8. Epub 2012/06/23. https://doi.org/10.1038/nature10831 PMID:22722205.

16. Baz M, Luke CJ, Cheng X, Jin H, Subbarao K. H5N1 vaccines in humans. Virus Res. 2013; 178(1):78–98. Epub 2013/06/04. https://doi.org/10.1016/j.virusres.2013.05.006 PMID: 23726847.

Broadly protective avian influenza vaccine

PLOS ONE | https://doi.org/10.1371/journal.pone.0186244 October 12, 2017 12 / 15

17. Nicholson KG, Colegate AE, Podda A, Stephenson I, Wood J, Ypma E, et al. Safety and antigenicity ofnon-adjuvanted and MF59-adjuvanted influenza A/Duck/Singapore/97 (H5N3) vaccine: a randomisedtrial of two potential vaccines against H5N1 influenza. Lancet. 2001; 357(9272):1937–43. Epub 2001/06/27. https://doi.org/10.1016/S0140-6736(00)05066-2 PMID: 11425416.

18. Stephenson I, Nicholson KG, Colegate A, Podda A, Wood J, Ypma E, et al. Boosting immunity to influ-enza H5N1 with MF59-adjuvanted H5N3 A/Duck/Singapore/97 vaccine in a primed human population.Vaccine. 2003; 21(15):1687–93. Epub 2003/03/18. PMID: 12639491.

19. Atmar RL, Keitel WA, Patel SM, Katz JM, She D, El Sahly H, et al. Safety and immunogenicity of nonad-juvanted and MF59-adjuvanted influenza A/H9N2 vaccine preparations. Clin Infect Dis. 2006; 43(9):1135–42. Epub 2006/10/10. https://doi.org/10.1086/508174 PMID: 17029131.

20. Cox RJ, Madhun AS, Hauge S, Sjursen H, Major D, Kuhne M, et al. A phase I clinical trial of a PER.C6®cell grown influenza H7 virus vaccine. Vaccine. 2009; 27:1889–97. https://doi.org/10.1016/j.vaccine.2009.01.116 PMID: 19368768.

21. Atmar RL, Keitel WA, Quarles JM, Cate TR, Patel SM, Nino D, et al. Evaluation of age-related differ-ences in the immunogenicity of a G9 H9N2 influenza vaccine. Vaccine. 2011; 29(45):8066–72. Epub2011/08/26. https://doi.org/10.1016/j.vaccine.2011.08.044 PMID: 21864622.

22. Vemula SV, Mittal SK. Production of adenovirus vectors and their use as a delivery system for influenzavaccines. Expert Opin Biol Ther. 2010; 10(10):1469–87. Epub 2010/09/09. https://doi.org/10.1517/14712598.2010.519332 PMID: 20822477.

23. Bangari DS, Mittal SK. Development of nonhuman adenoviruses as vaccine vectors. Vaccine. 2006; 24(7):849–62. Epub 2005/11/22. https://doi.org/10.1016/j.vaccine.2005.08.101 PMID: 16297508.

24. Zhu J, Huang X, Yang Y. Innate immune response to adenoviral vectors is mediated by both Toll-likereceptor-dependent and -independent pathways. J Virol. 2007; 81(7):3170–80. Epub 2007/01/19.https://doi.org/10.1128/JVI.02192-06 PMID: 17229689.

25. Sharma A, Tandon M, Ahi YS, Bangari DS, Vemulapalli R, Mittal SK. Evaluation of cross-reactive cell-mediated immune responses among human, bovine and porcine adenoviruses. Gene Ther. 2010; 17(5):634–42. https://doi.org/10.1038/gt.2010.1 PMID: 20164856.

26. Hoelscher M, Jayashankar L, Garg S, Veguilla V, Xlu, Singh N, et al. New pre-pandemic influenza vac-cines: an egg-and adjuvant-independent human adenoviral vector strategy induces long-lasting protec-tive immune responses in mice. Clin Pharmacol Ther. 2007; 82(6):665–71. https://doi.org/10.1038/sj.clpt.6100418 PMID: 17957181.

27. Gao W, Soloff AC, Lu X, Montecalvo A, Nguyen DC, Matsuoka Y, et al. Protection of mice and poultryfrom lethal H5N1 avian influenza virus through adenovirus-based immunization. J Virol. 2006; 80(4):1959–64. https://doi.org/10.1128/JVI.80.4.1959-1964.2006 PMID: 16439551.

28. Hoelscher MA, Garg S, Bangari DS, Belser JA, Lu X, Stephenson I, et al. Development of adenoviral-vector-based pandemic influenza vaccine against antigenically distinct human H5N1 strains in mice.Lancet. 2006; 367(9509):475–81. Epub 2006/02/14. https://doi.org/10.1016/S0140-6736(06)68076-8PMID: 16473124.

29. Van Kampen KR, Shi Z, Gao P, Zhang J, Foster KW, Chen DT, et al. Safety and immunogenicity of ade-novirus-vectored nasal and epicutaneous influenza vaccines in humans. Vaccine. 2005; 23(8):1029–36. Epub 2004/12/29. https://doi.org/10.1016/j.vaccine.2004.07.043 PMID: 15620476.

30. Gurwith M, Lock M, Taylor EM, Ishioka G, Alexander J, Mayall T, et al. Safety and immunogenicity of anoral, replicating adenovirus serotype 4 vector vaccine for H5N1 influenza: a randomised, double-blind,placebo-controlled, phase 1 study. Lancet Infect Dis. 2013; 13(3):238–50. Epub 2013/02/02. https://doi.org/10.1016/S1473-3099(12)70345-6 PMID: 23369412.

31. De Filette M, Martens W, Roose K, Deroo T, Vervalle F, Bentahir M, et al. An influenza A vaccine basedon tetrameric ectodomain of matrix protein 2. J Biol Chem. 2008; 283(17):11382–7. Epub 2008/02/07.https://doi.org/10.1074/jbc.M800650200 PMID: 18252707.

32. Fu TM, Grimm KM, Citron MP, Freed DC, Fan J, Keller PM, et al. Comparative immunogenicity evalua-tions of influenza A virus M2 peptide as recombinant virus like particle or conjugate vaccines in miceand monkeys. Vaccine. 2009; 27(9):1440–7. Epub 2009/01/17. https://doi.org/10.1016/j.vaccine.2008.12.034 PMID: 19146898.

33. Fan J, Liang X, Horton MS, Perry HC, Citron MP, Heidecker GJ, et al. Preclinical study of influenzavirus A M2 peptide conjugate vaccines in mice, ferrets, and rhesus monkeys. Vaccine. 2004; 22(23–24):2993–3003. Epub 2004/08/07. https://doi.org/10.1016/j.vaccine.2004.02.021 PMID: 15297047.

34. Ionescu RM, Przysiecki CT, Liang X, Garsky VM, Fan J, Wang B, et al. Pharmaceutical and immunolog-ical evaluation of human papillomavirus viruslike particle as an antigen carrier. J Pharm Sci. 2006; 95(1):70–9. Epub 2005/11/30. https://doi.org/10.1002/jps.20493 PMID: 16315228.

Broadly protective avian influenza vaccine

PLOS ONE | https://doi.org/10.1371/journal.pone.0186244 October 12, 2017 13 / 15

35. Schotsaert M, De Filette M, Fiers W, Saelens X. Universal M2 ectodomain-based influenza A vaccines:preclinical and clinical developments. Expert Rev Vaccines. 2009; 8(4):499–508. Epub 2009/04/08.https://doi.org/10.1586/erv.09.6 PMID: 19348565.

36. Steel J, Lowen AC, Wang TT, Yondola M, Gao Q, Haye K, et al. Influenza virus vaccine based on theconserved hemagglutinin stalk domain. MBio. 2010; 1(1). Epub 2010/08/07. https://doi.org/10.1128/mBio.00018-10 PMID: 20689752.

37. Bommakanti G, Citron MP, Hepler RW, Callahan C, Heidecker GJ, Najar TA, et al. Design of an HA2-based Escherichia coli expressed influenza immunogen that protects mice from pathogenic challenge.Proc Natl Acad Sci U S A. 2010; 107(31):13701–6. Epub 2010/07/10. https://doi.org/10.1073/pnas.1007465107 PMID: 20615991.

38. Krammer F, Pica N, Hai R, Tan GS, Palese P. Hemagglutinin Stalk-reactive antibodies are boosted fol-lowing sequential infection with seasonal and pandemic H1N1 influenza virus in mice. J Virol. 2012; 86(19):10302–7. Epub 2012/07/13. https://doi.org/10.1128/JVI.01336-12 PMID: 22787225.

39. Wang TT, Tan GS, Hai R, Pica N, Ngai L, Ekiert DC, et al. Vaccination with a synthetic peptide from theinfluenza virus hemagglutinin provides protection against distinct viral subtypes. Proc Natl Acad Sci U S A.2010; 107(44):18979–84. Epub 2010/10/20. https://doi.org/10.1073/pnas.1013387107 PMID: 20956293.

40. Yewdell JW, Bennink JR, Smith GL, Moss B. Influenza A virus nucleoprotein is a major target antigenfor cross-reactive anti-influenza A virus cytotoxic T lymphocytes. Proc Natl Acad Sci U S A. 1985; 82(6):1785–9. Epub 1985/03/01. PMID: 3872457.

41. Townsend AR, Skehel JJ, Taylor PM, Palese P. Recognition of influenza A virus nucleoprotein by an H-2-restricted cytotoxic T-cell clone. Virology. 1984; 133(2):456–9. Epub 1984/03/01. PMID: 6200990.

42. Townsend AR, Skehel JJ. The influenza A virus nucleoprotein gene controls the induction of both sub-type specific and cross-reactive cytotoxic T cells. J Exp Med. 1984; 160(2):552–63. Epub 1984/08/01.PMID: 6206181.

43. Ekiert DC, Bhabha G, Elsliger MA, Friesen RH, Jongeneelen M, Throsby M, et al. Antibody recognitionof a highly conserved influenza virus epitope. Science. 2009; 324(5924):246–51. Epub 2009/03/03.https://doi.org/10.1126/science.1171491 PMID: 19251591.

44. Kaminski DA, Lee FE. Antibodies against conserved antigens provide opportunities for reform in influ-enza vaccine design. Front Immunol. 2011; 2:76. Epub 2011/01/01. https://doi.org/10.3389/fimmu.2011.00076 PMID: 22566865.

45. Nabel GJ, Fauci AS. Induction of unnatural immunity: prospects for a broadly protective universal influ-enza vaccine. Nat Med. 2010; 16(12):1389–91. Epub 2010/12/08. https://doi.org/10.1038/nm1210-1389 PMID: 21135852.

46. Tan GS, Krammer F, Eggink D, Kongchanagul A, Moran TM, Palese P. A pan-H1 anti-hemagglutininmonoclonal antibody with potent broad-spectrum efficacy in vivo. J Virol. 2012; 86(11):6179–88. Epub2012/04/12. https://doi.org/10.1128/JVI.00469-12 PMID: 22491456.

47. Wei CJ, Boyington JC, McTamney PM, Kong WP, Pearce MB, Xu L, et al. Induction of broadly neutraliz-ing H1N1 influenza antibodies by vaccination. Science. 2010; 329(5995):1060–4. Epub 2010/07/22.https://doi.org/10.1126/science.1192517 PMID: 20647428.

48. van Olphen AL, Mittal SK. Development and characterization of bovine x human hybrid cell lines thatefficiently support the replication of both wild-type bovine and human adenoviruses and those with E1deleted. J Virol. 2002; 76(12):5882–92. Epub 2002/05/22. https://doi.org/10.1128/JVI.76.12.5882-5892.2002 PMID: 12021321.

49. Ng P, Parks RJ, Cummings DT, Evelegh CM, Sankar U, Graham FL. A high-efficiency Cre/loxP-basedsystem for construction of adenoviral vectors. Hum Gene Ther. 1999; 10(16):2667–72. Epub 1999/11/24. https://doi.org/10.1089/10430349950016708 PMID: 10566894.

50. Vemula SV, Ahi YS, Swaim AM, Katz JM, Donis R, Sambhara S, et al. Broadly protective adenovirus-based multivalent vaccines against highly pathogenic avian influenza viruses for pandemic prepared-ness. PloS One. 2013; 8(4):e62496. Epub 2013/05/03. https://doi.org/10.1371/journal.pone.0062496PMID: 23638099.

51. Hoelscher MA, Singh N, Garg S, Jayashankar L, Veguilla V, Pandey A, et al. A broadly protective vac-cine against globally dispersed clade 1 and clade 2 H5N1 influenza viruses. J Infect Dis. 2008; 197(8):1185–8. Epub 2008/05/09. https://doi.org/10.1086/529522 PMID: 18462165.

52. Noblitt LW, Bangari DS, Shukla S, Knapp DW, Mohammed S, Kinch MS, et al. Decreased tumorigenicpotential of EphA2-overexpressing breast cancer cells following treatment with adenoviral vectors thatexpress EphrinA1. Cancer Gene Ther. 2004; 11(11):757–66. Epub 2004/09/11. https://doi.org/10.1038/sj.cgt.7700761 PMID: 15359289.

53. Pandey A, Singh N, Vemula SV, Couetil L, Katz JM, Donis R, et al. Impact of preexisting adenovirusvector immunity on immunogenicity and protection conferred with an adenovirus-based H5N1 influenzavaccine. PloS One. 2012; 7:e33428. https://doi.org/10.1371/journal.pone.0033428 PMID: 22432020.

Broadly protective avian influenza vaccine

PLOS ONE | https://doi.org/10.1371/journal.pone.0186244 October 12, 2017 14 / 15

54. Reed LJ, Muench H. A simple method of estimating fifty per cent endpoints. Am J Epidemiol. 1938;27:493–7.

55. Mittal SK, Middleton DM, Tikoo SK, Babiuk LA. Pathogenesis and immunogenicity of bovine adenovirustype 3 in cotton rats (Sigmodon hispidus). Virology. 1995; 213(1):131–9. Epub 1995/10/20. https://doi.org/10.1006/viro.1995.1553 PMID: 7483256.

56. Cao W, Liepkalns JS, Hassan AO, Kamal RP, Hofstetter AR, Amoah S, et al. A highly immunogenicvaccine against A/H7N9 influenza virus. Vaccine. 2016; 34(6):744–9. Epub 2016/01/15. https://doi.org/10.1016/j.vaccine.2015.12.062 PMID: 26765287.

57. Thomas PG, Keating R, Hulse-Post DJ, Doherty PC. Cell-mediated protection in influenza infection.Emerg Infect Dis. 2006; 12(1):48–54. Epub 2006/02/24. https://doi.org/10.3201/eid1201.051237 PMID:16494717.

58. Fiers W, De Filette M, Birkett A, Neirynck S, Min Jou W. A "universal" human influenza A vaccine. VirusRes. 2004; 103(1–2):173–6. Epub 2004/05/28. https://doi.org/10.1016/j.virusres.2004.02.030 PMID:15163506.

59. Assarsson E, Bui HH, Sidney J, Zhang Q, Glenn J, Oseroff C, et al. Immunomic analysis of the reper-toire of T-cell specificities for influenza A virus in humans. J Virol. 2008; 82(24):12241–51. Epub 2008/10/10. https://doi.org/10.1128/JVI.01563-08 PMID: 18842709.

60. Grant E, Wu C, Chan KF, Eckle S, Bharadwaj M, Zou QM, et al. Nucleoprotein of influenza A virus is amajor target of immunodominant CD8+ T-cell responses. Immunol Cell Biol. 2013; 91(2):184–94. Epub2013/02/13. https://doi.org/10.1038/icb.2012.78 PMID: 23399741.

61. Medina RA, Garcia-Sastre A. Influenza A viruses: new research developments. Nat Rev Microbiol.2011; 9(8):590–603. Epub 2011/07/13. https://doi.org/10.1038/nrmicro2613 PMID: 21747392.

62. Du L, Zhou Y, Jiang S. Research and development of universal influenza vaccines. Microbes Infect.2010; 12(4):280–6. Epub 2010/01/19. https://doi.org/10.1016/j.micinf.2010.01.001 PMID: 20079871.

63. Atassi MZ, Webster RG. Localization, synthesis, and activity of an antigenic site on influenza virus hem-agglutinin. Proc Natl Acad Sci U S A. 1983; 80(3):840–4. Epub 1983/02/01. PMID: 6187004.

64. Stanekova Z, Kiraly J, Stropkovska A, Mikuskova T, Mucha V, Kostolansky F, et al. Heterosubtypic pro-tective immunity against influenza A virus induced by fusion peptide of the hemagglutinin in comparisonto ectodomain of M2 protein. Acta Virol. 2011; 55(1):61–7. Epub 2011/03/26. PMID: 21434706.

65. Stanekova Z, Mucha V, Sladkova T, Blaskovicova H, Kostolansky F, Vareckova E. Epitope specificity ofanti-HA2 antibodies induced in humans during influenza infection. Influenza Other Respir Viruses. 2012;6(6):389–95. Epub 2012/01/13. https://doi.org/10.1111/j.1750-2659.2011.00328.x PMID: 22236105.

66. Lee JS, Chowdhury MY, Moon HJ, Choi YK, Talactac MR, Kim JH, et al. The highly conserved HA2 pro-tein of the influenza A virus induces a cross protective immune response. J Virol Methods. 2013; 194(1–2):280–8. Epub 2013/09/06. https://doi.org/10.1016/j.jviromet.2013.08.022 PMID: 24004822.

67. Epstein SL, Kong WP, Misplon JA, Lo CY, Tumpey TM, Xu L, et al. Protection against multiple influenzaA subtypes by vaccination with highly conserved nucleoprotein. Vaccine. 2005; 23(46–47):5404–10.Epub 2005/07/14. https://doi.org/10.1016/j.vaccine.2005.04.047 PMID: 16011865.

68. Wesley RD, Tang M, Lager KM. Protection of weaned pigs by vaccination with human adenovirus 5recombinant viruses expressing the hemagglutinin and the nucleoprotein of H3N2 swine influenzavirus. Vaccine. 2004; 22(25–26):3427–34. Epub 2004/08/17. https://doi.org/10.1016/j.vaccine.2004.02.040 PMID: 15308368.

69. Chen W, Anton LC, Bennink JR, Yewdell JW. Dissecting the multifactorial causes of immunodominancein class I-restricted T cell responses to viruses. Immunity. 2000; 12(1):83–93. Epub 2000/02/08. PMID:10661408.

70. Ulmer JB, Donnelly JJ, Parker SE, Rhodes GH, Felgner PL, Dwarki VJ, et al. Heterologous protectionagainst influenza by injection of DNA encoding a viral protein. Science. 1993; 259(5102):1745–9. Epub1993/03/19. PMID: 8456302.

71. Tao P, Luo M, Pan R, Ling D, Zhou S, Tien P, et al. Enhanced protective immunity against H5N1 influ-enza virus challenge by vaccination with DNA expressing a chimeric hemagglutinin in combination withan MHC class I-restricted epitope of nucleoprotein in mice. Antiviral Res. 2009; 81(3):253–60. Epub2009/01/13. https://doi.org/10.1016/j.antiviral.2008.12.009 PMID: 19135483.

72. Deliyannis G, Jackson DC, Ede NJ, Zeng W, Hourdakis I, Sakabetis E, et al. Induction of long-termmemory CD8(+) T cells for recall of viral clearing responses against influenza virus. J Virol. 2002; 76(9):4212–21. Epub 2002/04/05. https://doi.org/10.1128/JVI.76.9.4212-4221.2002 PMID: 11932386.

73. Crowe SR, Turner SJ, Miller SC, Roberts AD, Rappolo RA, Doherty PC, et al. Differential antigen pre-sentation regulates the changing patterns of CD8+ T cell immunodominance in primary and secondaryinfluenza virus infections. J Exp Med. 2003; 198(3):399–410. Epub 2003/07/30. https://doi.org/10.1084/jem.20022151 PMID: 12885871.

Broadly protective avian influenza vaccine

PLOS ONE | https://doi.org/10.1371/journal.pone.0186244 October 12, 2017 15 / 15